The proportion of all AIDS cases reported among women has more than tripled since 1985. It appears that relationships with men are the greatest source of HIV risk for women. Women's ability to successfully negotiate is an important component of the process. Although the ultimate decision to use preventative measures is the man's, in a consensual encounter, there is still an opportunity for a woman to insist or influence that decision. Alcohol consumption can interfere with the effectiveness of a woman's negotiation and the decision not to be involved in risky behavior. We propose that these outcomes are cognitively mediated by a series of primary and secondary appraisals during these encounters. Primary appraisals are those directed toward recognizing that a situation is likely to lead to these inevitable men-women encounters. Secondary appraisals involve assessing the advantages and disadvantages of engaging in negotiation, and can influence the effectiveness of the negotiation. In addition to being influenced by alcohol, both directly through actual consumption and indirectly through expectancies, these secondary appraisals and subsequent negotiation can be influenced by the type of relationship the woman has with the man and her knowledge about his relationship history. Ultimately these will lead to her decision whether to have unprotected relationship. Theoretical models of cognitive mediation and the effects of alcohol consumption and expectancies provide the foundation for this work. Four experiments are proposed to investigate physiological and expectancy set effects in young women of drinking age. An additional four experiments will investigate the role of alcohol expectancies in 18-20 year old women. Manipulated variables in the between-subjects experimental design include relationship type and knowledge of partner's relationship history, depending on the experiment. The role of background factors related to the relational self, and victimization history will also be examined. This work will provide important information about how women's alcohol consumption and expectancies about relationships affect their assessment of situations, negotiation, and decisions about unprotected involvement. Findings have implications for designing prevention interventions, especially related to alcohol's role as a risk factor for contracting HIV.